18 replies to this topic

[FSQAManager2025]

Good morning! This is going to sound dumb, but I'd rather ask a dumb question than try to assume something on my own. When it comes to interpreting test results, what are the limits? My facility receives our E. coli results using ECC (cfu / cm²) for the result. Most of the time it is less than <0.03. Last week it was .6. What is considered a failure? What is the limit for this? I was told by my predecessor that the limit is 1,000,000, but this seems like an extremely high number. I'd just like to verify. Thank you! 

[kconf]

It depends on some things. Is this an env sample you are talking about or food sample?

 

What is the nature of food? Raw/finished/produce? What type?

 

If you are being reported <0.03 it is most likely MPN, not CFU. CFU is usually a whole number. Maybe your lab changed the method this time. I'd ask them directly. 

 

The limit is what you set based on what it is and your process. I've seen millions of colonies (coli) on raw meat shelf life. It is not unusual. 

[BigGaz1982]

Limits are based very much on the product, and it's intended use.

 

The Health Protection Agency have released guidelines which can be useful when determining limits. This can be found here

 

Also, the limits of detection from our provider come into play too.

 

For example, a result of <0.03 could mean that the limit of detection is 0.03 for your provider, and therefore it is not detected. If the lab has changed or outsourced, then the 0.6 could also be the limit of detection.

 

Usually the cert would also state "Not Detected" too, but you can clarify that with the provider.

 

If the limit of detection (LoD) is 0.03, and this time the result has come as 0.6, then this would potentially indicate that E.Coli has been detected.

 

Also, the limit for E.Coli is generally "Not Detected". Any result which shows the presence of E.Coli would be a failure. The risk then comes down to the type of foodstuff and thus the source of the E.Coli. But any E.Coli detection would be a failure.

Edited by BigGaz1982, Today, 04:26 PM.

[GMO]

How can a limit of detection be anything but a whole number or less than a whole number? If there are dilution effects then it might be <10 or <20 but <0.03 makes no sense to me.

But I'm not a micro expert.

I'm assuming a swab is meant as it's per cm2. 

[kconf]

I think - not a limit of detection, the result itself recorded as <0.03 is what the OP means. 

[BigGaz1982]

I think - not a limit of detection, the result itself recorded as <0.03 is what the OP means. 

<0.03 would indicate that the lowest the lab is able to quantifiably detected is 0.03. Anything lower than this, they cannot detect. So, as far as they are concerned it is Not Present, but they cannot confirm with absolute certainty that the level of E.Coli is zero due to the accuracy of their tests.

 

In the instance of a result reading <0.03 this would be considered a successful sample. If the same lab is used and records a level of 0.6, then this is a confirmation that there IS E.Coli present and they can actively determine the exact level in the test sample.

[FSQAManager2025]

It is a generic e. coli sample. A sponge used on a carcass, specifically raw beef. 

It depends on some things. Is this an env sample you are talking about or food sample?

 

What is the nature of food? Raw/finished/produce? What type?

 

If you are being reported <0.03 it is most likely MPN, not CFU. CFU is usually a whole number. Maybe your lab changed the method this time. I'd ask them directly. 

 

The limit is what you set based on what it is and your process. I've seen millions of colonies (coli) on raw meat shelf life. It is not unusual. 

[kconf]

Oh I see. Then there is a special calculation for carcass. When doing Pre and Post, pres could be that high. 

[FSQAManager2025]

Your link is for 'ready-to-eat' foods placed on the market. My facilities products are raw beef. Would this still apply? 

Limits are based very much on the product, and it's intended use.

 

The Health Protection Agency have released guidelines which can be useful when determining limits. This can be found here

 

Also, the limits of detection from our provider come into play too.

 

For example, a result of <0.03 could mean that the limit of detection is 0.03 for your provider, and therefore it is not detected. If the lab has changed or outsourced, then the 0.6 could also be the limit of detection.

 

Usually the cert would also state "Not Detected" too, but you can clarify that with the provider.

 

If the limit of detection (LoD) is 0.03, and this time the result has come as 0.6, then this would potentially indicate that E.Coli has been detected.

 

Also, the limit for E.Coli is generally "Not Detected". Any result which shows the presence of E.Coli would be a failure. The risk then comes down to the type of foodstuff and thus the source of the E.Coli. But any E.Coli detection would be a failure.

[Scampi]

Helpful to know where in the world you are but below should help

 

Table 3. Values for Marginal and Unacceptable Results for E. coli performance criteria Slaughter Acceptable Range Marginal Range Unacceptable Range Class Cattle *negative positive above 100 but not above 100 cfu/cm2 cfu/cm2 Swine 10 cfu/cm2 above 10 cfu/cm2 above 10,000 but not above 10,000 cfu/cm2 cfu/cm2 * An excised sample is considered negative when no E. coli colonies are present on plate(s) Of the lowest dilution (10- 1 1). If E. coli colonies are present, multiply the average plate count by the appropriate dilution factor and record the result as cfu/cm2 .

https://www.fsis.usd...Swine_Estab.pdf

[FSQAManager2025]

I am in the United States. Thank you! 

Helpful to know where in the world you are but below should help

 

Table 3. Values for Marginal and Unacceptable Results for E. coli performance criteria Slaughter Acceptable Range Marginal Range Unacceptable Range Class Cattle *negative positive above 100 but not above 100 cfu/cm2 cfu/cm2 Swine 10 cfu/cm2 above 10 cfu/cm2 above 10,000 but not above 10,000 cfu/cm2 cfu/cm2 * An excised sample is considered negative when no E. coli colonies are present on plate(s) Of the lowest dilution (10- 1 1). If E. coli colonies are present, multiply the average plate count by the appropriate dilution factor and record the result as cfu/cm2 .

https://www.fsis.usd...Swine_Estab.pdf

[SHQuality]

CFU/cm³ = CFU/ml

 

CFU stands for colony forming units. I would be extremely surprised if the lab measured E. coli colonies in anything other than whole colonies even when multiple dilution steps are involved. Either it's MPN or it's something else.

 

I'm surprised you are asking about the limits for this test, because any lab worth its salt would have that limit listed on the report. 1 million seems extremely large to me as well. I would expect less than 100 or less than 1000 at upper levels of acceptable depending on the type of product we're talking about.

 

You definitely need to talk to the lab as to why the agreed upon limits for your product are not on the report.

[kconf]

OP states it is ECC petrifilm and the high numbers are likely for coliform.

 

Carcass swabbing CFU can be in fraction, doesn't have to be a whole number. When nothing (0 colonies) is detected on film, it is <0.03. If OP has 0.6 reported, it means there were 20 colonies grown on film. 

[FSQAManager2025]

 I have attached a copy of the report. We are testing raw beef carcasses. 

CFU/cm³ = CFU/ml

 

CFU stands for colony forming units. I would be extremely surprised if the lab measured E. coli colonies in anything other than whole colonies even when multiple dilution steps are involved. Either it's MPN or it's something else.

 

I'm surprised you are asking about the limits for this test, because any lab worth its salt would have that limit listed on the report. 1 million seems extremely large to me as well. I would expect less than 100 or less than 1000 at upper levels of acceptable depending on the type of product we're talking about.

 

You definitely need to talk to the lab as to why the agreed upon limits for your product are not on the report.

 

[FSQAManager2025]

What would my critical limit be? 

OP states it is ECC petrifilm and the high numbers are likely for coliform.

 

Carcass swabbing CFU can be in fraction, doesn't have to be a whole number. When nothing (0 colonies) is detected on film, it is <0.03. If OP has 0.6 reported, it means there were 20 colonies grown on film. 

[kconf]

No Ecoli

[kconf]

https://www.fsis.usd...Swine_Estab.pdf

 

Read this. 

[SHQuality]

What would my critical limit be? 

 

No Ecoli

So a report of 0.6 when <0,03 is expected is over the limit. Any idea why this result came back like that, FSQA?

 

kconf, thanks for clarifying. I haven't worked in the meat industry myself. Your instinct turned out to be right.

Can you clarify if it is common for the limit to not be on the report contrary to what I would expect from other industries?

Edited by SHQuality, Today, 07:46 PM.

[kconf]

Yes, it is common to encounter. I have never worked in meat industry myself, but I have experience in doing such tests in lab setting. 

 

OP would have a better idea on why it would occur and what actions to take. E.coli found on carcass swab petri could be a harmless strain, and not the 0157:H7, otherwise they would be doing a more specific test such as PCR. If I knew the process, I would have been able to answer better.